alinaqi/claude-bootstrap
63 skills28.2k installs44.5k starsGitHub
Install
npx skills add https://github.com/alinaqi/claude-bootstrapSkills in this repo
1Android Kotlinandroid-kotlin is an agent skill for Android development using Kotlin coroutines, Jetpack Compose, Hilt dependency injection, Room persistence, and MockK testing conventions. It documents a layered project layout with data, domain, and ui packages, Gradle Kotlin DSL setup with Compose BOM, Hilt KSP processors, Room, and coroutines-test plus Turbine for Flow testing. ViewModel examples use HiltViewModel, MutableStateFlow uiState, viewModelScope launches, and SavedStateHandle arguments. Repository patterns emit cached Room data first then refresh from Retrofit or Ktor remote services on Dispatchers.IO. Testing guidance covers MockK, kotlinx-coroutines-test, and Compose ui-test-junit4 instrumentation. The skill auto-applies on Kotlin and Gradle paths listed in metadata when working inside Android modules. Developers reach for it when scaffolding Compose screens, wiring Hilt modules, implementing Flow-based repositories, or aligning Gradle dependencies with current Android toolchain versions across app modules.2.9kinstalls2Pwa Developmentpwa-development guides Progressive Web App implementation across HTTPS, service workers, and web app manifests. It documents installability criteria for Chrome, required manifest fields including 192px and 512px icons plus maskable icons, and enhanced manifest features such as shortcuts, share targets, and file handlers. Service worker patterns cover install, activate, and fetch lifecycle with cache-first, network-first, and stale-while-revalidate strategies per resource type. Workbox setup includes Vite PWA plugin configuration, precaching, runtime caching with ExpirationPlugin, and background sync for offline form queues. The skill also covers offline.html fallbacks, online and offline detection, install prompt handling, push notifications, performance optimizations, Lighthouse audits, common mistakes like missing maskable icons or stale caches, project structure under public/icons, and framework guides for Next.js next-pwa, Create React App PWA template, and vite-plugin-pwa with manual service worker examples.2.7kinstalls3Playwright TestingPlaywright E2E Testing covers installation, playwright.config.ts setup, Page Object Model structure, locator strategy, authentication fixtures, and CI patterns for web apps. Setup uses npm init playwright or @playwright/test install with projects for chromium, firefox, webkit, and mobile viewports plus a setup dependency for global auth. Configuration enables fullyParallel runs, CI retries, trace on first retry, screenshots and video on failure, and webServer dev bootstrapping. Project structure places fixtures, pages, tests, utils, and auth.setup.ts under e2e with gitignored .auth storage state. Locator priority favors getByRole, getByLabel, getByTestId, and avoids brittle CSS or XPath selectors. Page objects extend BasePage with navigate, waitForPageLoad, and role-based elements like LoginPage and DashboardPage. Global auth setup logs in once and saves storageState for dependent projects. Chaining locators narrows forms and product cards. Sources reference Playwright best practices and Better Stack guidance for reliable cross-browser automation.2.6kinstalls4Code Reviewcode-review is an agent skill from alinaqi/claude-bootstrap that mandatory code reviews via /code-review before commits and deploys. # Code Review Skill **Purpose:** Enforce automated code reviews as a mandatory guardrail before every commit and deployment. Choose between Claude, OpenAI Codex, Google Gemini, or multiple engines for comprehensive analysis. **Sub-skills:** - [adr-gate.md](./adr-gate.md) — Pre-review ADR and spec enforcement --- ## Pre-Review: ADR Gate (Mandat Developers invoke code-review during build/integrations work for ai & agent building tasks. The skill documents triggers, prerequisites, and step-by-step workflows grounded in SKILL.md. Compatible with Claude Code, Cursor, and Codex agent runtimes that load marketplace skills. Review the Security Audits panel on this listing before installing in production environments. Category AI & Agent Building with ai vertical focus supports repeatable agent-guided delivery.1.3kinstalls5Ui Mobileui-mobile is a mobile UI design skill from alinaqi/claude-bootstrap that enforces non-negotiable accessibility standards while generating React Native, iOS, and Android components. The skill mandates a minimum 44x44-point touch size for every interactive element, applies StyleSheet patterns for buttons and links, and scopes to tsx, jsx, ios, android, and dart file paths. Developers reach for ui-mobile when building new mobile screens, refactoring touch targets, or auditing components for WCAG-aligned mobile accessibility before release. The skill is user-invocable false and marked medium effort, signaling pattern-heavy UI work rather than one-off fixes.1.1kinstalls6Posthog Analyticsposthog-analytics is an alinaqi/claude-bootstrap skill for implementing product analytics with PostHog, drawing on official PostHog documentation for product analytics and feature flags. The skill follows a measure-what-matters philosophy—tracking events that answer specific product questions rather than logging everything. Developers reach for posthog-analytics when adding analytics, feature flags, or event tracking with PostHog SDKs and configuring project-specific dashboards. It covers user identification, event schema decisions, and feature-flag rollout patterns so analytics code ships correctly the first time. Medium effort and non-user-invocable by default, it is meant for agent-driven implementation during active product development.760installs7Flutterflutter is a skill from alinaqi/claude-bootstrap that enforces Flutter development conventions for Riverpod state management, Freezed immutable models, go_router navigation, and mocktail testing. It auto-applies to paths including **/*.dart, pubspec.yaml, lib/**, and test/** with medium effort. The documented structure places core utilities under lib/core with router and theme modules plus data and feature layers. Developers reach for it when editing Dart code in a bootstrap-scaffolded app and want agents to follow consistent patterns instead of mixing state-management styles.696installs8Supabase Nextjssupabase-nextjs is an alinaqi/claude-bootstrap skill that teaches agents how to build Next.js applications with Supabase and Drizzle ORM using production-oriented patterns. Its core principle is Drizzle for queries, Supabase for auth and storage, and server components by default within the App Router. The skill scopes paths across src/app/**, src/db/**, and supabase/** so agents know where auth, database, and migration files belong. It references official Supabase Next.js server-side auth guidance and Drizzle plus Supabase database documentation as authoritative sources. Developers reach for supabase-nextjs when bootstrapping SaaS or API-backed Next.js projects that need typed SQL access without abandoning Supabase-managed auth, file storage, and realtime channels. Effort is rated medium, reflecting integrated setup across ORM schemas, Supabase clients, and server component data fetching rather than a single-file tutorial.666installs9Supabasesupabase is an alinaqi claude-bootstrap skill for core Supabase CLI workflows across database, auth, storage, and Edge Functions with a local-first, migrations-in-version-control principle—never touch production directly. The documented stack spans six services: PostgreSQL database with extensions, Auth with OAuth providers, Storage with RLS, Edge Functions on Deno, Realtime WebSocket subscriptions, and Vector embeddings via pgvector. Setup flows cover brew or npm Supabase CLI install, supabase init creating config.toml, seed.sql, and migrations/, supabase link to remote project refs, supabase db pull for schema sync, and supabase start for a local API on port 54321. The skill auto-activates on supabase/** paths, supabase.* files, and .env patterns with medium effort and is not user-invocable by default. Developers reach for it when adding RLS policies, writing SQL migrations, configuring auth or storage buckets, or deploying Edge Functions through CI/CD instead of dashboard edits.603installs10Android Javaandroid-java is a Claude Bootstrap skill from alinaqi/claude-bootstrap for Android Java development using MVVM architecture, ViewBinding, and Espresso testing. The skill activates on paths including **/*.java, android/**, and **/build.gradle and prescribes a layered project structure with data, domain, and DI modules covering Room, Retrofit, Hilt or Dagger, and repository patterns. Developers reach for android-java when modifying Java Android apps and need consistent MVVM scaffolding, layout binding conventions, and test-oriented structure instead of ad-hoc Activity code. The skill is marked medium effort and is not user-invocable, firing automatically when matching Android Java files are in scope.599installs11React Nativereact-native is a mobile development skill from alinaqi/claude-bootstrap that steers agents toward a `src/core`, `src/components`, and `src/screens` layout with barrel exports, co-located `Button.test.tsx` files, and separation of pure business logic from React UI. It activates on edits to `.tsx`, `.jsx`, `ios/**`, `android/**`, and `app.json` paths with medium effort. Developers reach for react-native when scaffolding or refactoring React Native apps and need consistent screen components, reusable UI modules, and typed functional patterns instead of ad-hoc folder sprawl. The skill is user-invocable false, so agents load it automatically during mobile file edits.533installs12User Journeysuser-journeys is a medium-effort skill from alinaqi/claude-bootstrap focused on user experience flows: journey mapping, UX validation, and error recovery design. It captures how users actually navigate, emotional states at each step such as frustration or confusion, and recovery from mistakes rather than idealized spec paths. The philosophy states specs test features while journeys test experiences—a feature can pass specs yet deliver poor UX. Developers invoke it when mapping flows, validating UX, or designing error recovery before or during prototype validation.497installs13Aws Dynamodbaws-dynamodb is a Claude Bootstrap skill from alinaqi/claude-bootstrap focused on AWS DynamoDB single-table design, global secondary index patterns, and AWS SDK v3 client usage in TypeScript or Python. The skill auto-triggers on paths matching dynamodb, dynamo, serverless configs, and template.yaml, signaling serverless and SAM-oriented projects. It references official DynamoDB documentation, SDK v3 guides, and AWS best practices for modeling access patterns before table creation. Developers reach for aws-dynamodb when designing keys, GSIs, or query paths for single-digit millisecond NoSQL workloads at scale. The README emphasizes access-pattern-first modeling rather than relational table-per-entity designs, making it suited to Lambda, API Gateway, and event-driven backends.490installs14Code Deduplicationcode-deduplication is an alinaqi/claude-bootstrap agent skill that stops semantic code bloat by enforcing a check-before-write discipline for AI-generated utilities. Agents maintain CODE_INDEX.md at the project root, organized by capability with function location, purpose, and parameters, then search the index and codebase before creating new helpers. The skill defines file-header and docstring templates, a four-step pre-write decision tree, six-category periodic audit checklists, and optional ChromaDB or LanceDB vector search thresholds for codebases above 200 files. Commands /update-code-index and /audit-duplicates keep the registry current. Developers reach for code-deduplication when agents keep reimplementing date formatters, validators, or API clients that already exist under different names.450installs15Klaviyoklaviyo is an alinaqi/claude-bootstrap skill for integrating Klaviyo email and SMS marketing into ecommerce and SaaS products. It covers server-side API calls to https://a.klaviyo.com/api with revision 2024-10-15 headers, klaviyo-api SDK setup in Node.js and Python, profile upserts, server-side event tracking for Viewed Product through Placed Order, list and segment management, flow triggers, and signed webhook verification. Rate-limit guidance documents 75 requests per second burst and 700 requests per minute steady with retry-after handling. Integration patterns include order sync after checkout and subscription status updates with revenue attribution via event value fields. The skill includes environment variable checklists and anti-patterns such as missing Items arrays on order events. Reach for klaviyo when wiring onboarding drips, abandoned-cart flows, or storefront event pipelines to Klaviyo rather than building a custom ESP integration from scratch.410installs16Aeo Optimizationaeo-optimization is an AI Engine Optimization skill from alinaqi/claude-bootstrap that structures portfolio and marketing pages for citation by ChatGPT, Claude, Perplexity, and Google AI Overviews. Based on HubSpot’s AEO guide and industry practice, it applies semantic triples, reusable page templates, and topical content clusters so AI engines can quote self-contained answers about a brand. Developers reach for it when documentation sites, product landing pages, or developer portfolios underperform in AI-generated answers despite solid traditional SEO. The skill guides entity mapping, rewrite patterns, and visibility tracking rather than code deployment. Effort is marked medium in the manifest. Use it while preparing launch content or refreshing docs you want LLM crawlers and answer engines to reference accurately.396installs17Nodejs Backendnodejs-backend is an alinaqi claude-bootstrap skill encoding Node.js backend conventions for Express and Fastify projects. It organizes code into src/core for pure domain logic—types, errors, and services—and src/infra for side effects like repositories and external IO. The skill activates on paths including src/api, src/routes, src/server, src/middleware, server, and api directories, guiding agents through route handlers, middleware chains, auth integration, and server bootstrap suitable for SaaS and agent backends. Developers reach for nodejs-backend when scaffolding new API services or aligning existing Node code with repository and domain separation. It marks medium effort and is not user-invocable directly, instead triggering automatically when agents edit matching backend files.387installs18Securitysecurity is an agent skill from alinaqi/claude-bootstrap that enforces OWASP security patterns, secrets management, and automated security testing across web and API projects. The skill treats security as non-optional: assume all input is malicious, all secrets will leak if committed, and all dependencies have vulnerabilities. Required setup covers `.gitignore` rules for secrets, dependency scanning, auth hardening, input validation, and secure defaults on web and API surfaces. The manifest grants Read, Glob, Grep, and Bash tools with high effort classification for thorough review. Developers invoke security when writing auth flows, handling user input or API keys, or when an explicit security review is requested. The skill fits pre-merge and pre-release gates rather than one-time audits, integrating security checks into ongoing development workflows.379installs19Agentic Developmentagentic-development is a high-effort agent skill from alinaqi/claude-bootstrap for building autonomous tool-using LLM systems. It defaults to Pydantic AI in Python for type-safe structured outputs and the Claude Agent SDK in Node.js for official Anthropic tool loops with streaming. The skill encodes Explore-Plan-Execute-Verify workflows, three-component agent architecture (model, tools, instructions), multi-agent handoff patterns, memory layers, guardrails, model selection guidance, and testing strategies for tools and agent behavior. Developers reach for agentic-development when standing up orchestrators, specialized sub-agents, tool registries, or repository conventions for Claude Code-style autonomous tasks. Sources reference Anthropic Agent SDK docs, Pydantic AI, OpenAI agent tracks, and Gemini agent guidance in one consolidated playbook.368installs20Ui Webui-web is an agent skill from alinaqi/claude-bootstrap for building and styling web UI in TSX, JSX, CSS, SCSS, and Tailwind config files. It enforces non-negotiable WCAG 2.1 AA checks including 4.5:1 normal text contrast, 3:1 large text and UI component contrast, and forbidden low-contrast pairings such as gray-400 on white. The skill covers glassmorphism treatments, dark mode patterns, typography, spacing, component composition, and accessible focus indicators while matching bootstrap design tokens. Agents apply it when editing paths like `**/*.tsx`, `**/*.jsx`, `**/*.css`, and `tailwind.config.*`. Reach for ui-web when scaffolding dashboards, marketing pages, or design-system-aligned components where accessibility failures would block release.355installs21Woocommercewoocommerce is a claude-bootstrap agent skill for integrating with WooCommerce stores through the wc/v3 REST API at `/wp-json/wc/v3/`. It documents prerequisites—HTTPS, non-Plain WordPress permalinks, and REST API key generation—and provides Node.js (`@woocommerce/woocommerce-rest-api`) and Python (`woocommerce`) client patterns for products, variable products, orders, customers, coupons, categories, reports, and webhooks. Webhook handlers include HMAC-SHA256 signature verification via `x-wc-webhook-signature` for topics like order.created and product.updated. Pagination uses `X-WP-Total` and `X-WP-TotalPages` headers for bulk catalog sync, and error handling covers 401 credential failures and 429 rate-limit retries. Developers reach for woocommerce when building order sync pipelines, inventory integrations, custom checkout flows, or webhook-driven fulfillment automation against live WooCommerce stores.353installs22Ui Testingui-testing is an alinaqi/claude-bootstrap skill (effort: medium, user-invocable: false) triggered on paths `**/*.test.tsx`, `**/*.spec.tsx`, and `**/*.stories.*`. It loads companion guides ui-web.md or ui-mobile.md and provides a pre-flight checklist covering visibility (4.5:1 contrast), 44px minimum touch targets, hover/press/focus/disabled/loading states, dark-mode readability, and 320px responsive checks. It documents Tailwind safe color pairs, common JSX fixes for invisible buttons and missing focus rings, optional eslint-plugin-jsx-a11y setup, and a Playwright @axe-core/playwright example. Developers reach for this skill after generating new UI components when they need fast accessibility verification rather than a full visual regression suite.327installs23React Webreact-web in alinaqi/claude-bootstrap guides React web development for .tsx and .jsx files under src/components, src/pages, and src/app. The skill mandates test-first development: write Component.test.tsx, confirm failing tests, implement minimal passing code, then refactor. It documents project structure separating core business logic from UI, functional components with explicit props interfaces, and hooks like useHome that keep pages presentational. State guidance covers local useState, Zustand global stores, and TanStack React Query for server state with staleTime and invalidation patterns. Routing uses React Router with protected routes; forms pair React Hook Form with Zod schemas. Styling prefers CSS modules with Tailwind as an alternative. Testing covers React Testing Library unit tests, hook tests with renderHook, and Playwright E2E specs. Performance patterns include memo, useCallback, useMemo, and lazy route splitting. Developers reach for react-web when adding bootstrap-stack React pages, enforcing TFD on new components, or wiring React Query and Zustand in Vite projects.305installs24Commit Hygienecommit-hygiene is a git discipline skill from alinaqi/claude-bootstrap that keeps bootstrap repositories readable by enforcing one logical change per commit, PR size limits, commit thresholds, and stacked PR patterns. The skill advises when to commit before a change set grows too large and promotes bisect-friendly history suitable for team review. Developers reach for commit-hygiene while committing code, opening PRs, or when diffs are expanding beyond comfortable review size. The philosophy centers on atomic commits where each commit stands alone as a single logical unit, making rollbacks and code archaeology straightforward during ship and review cycles.304installs25Site Architecturesite-architecture is a technical site structure skill from alinaqi/claude-bootstrap that plans information architecture, routes, layouts, and content zones before implementation so marketing sites and app shells stay coherent and SEO-friendly. The skill configures technical discovery elements including robots.txt, sitemap files, meta tags, and Core Web Vitals considerations, explicitly targeting crawlers such as GPTBot, ClaudeBot, and PerplexityBot alongside Google and Bing. Path triggers include robots.txt, sitemap files, HTML pages, and public assets. Developers reach for site-architecture when setting up a new marketing site, app shell, or technical SEO foundation rather than tuning individual blog posts or writing application business logic.302installs26Web Contentweb-content is a Claude Code skill from alinaqi/claude-bootstrap for producing on-site copy optimized for both traditional SEO and generative-engine optimization. The skill provides page-type templates for homepages, product pages, blog posts, FAQ pages, and landing pages, each with required schema markup (Organization, Article, FAQPage, Product, HowTo). It defines AI-friendly content formats—TL;DR boxes, definition blocks, comparison tables, stat boxes with sources, and step-by-step guides—plus platform-specific guidance for ChatGPT, Perplexity, Claude, and Gemini. A dual SEO+GEO checklist covers title tags, meta descriptions, visible publication dates, E-E-A-T signals, and GA4 regex filters for AI referral tracking. The philosophy is SEO gets clicks while GEO gets citations; the skill cites that AI traffic grew 1,200% between July 2024 and February 2025. Developers reach for web-content when scaffolding new pages, blog posts, or marketing copy that must rank in Google and surface in AI-generated answers.294installs27Codex Reviewcodex-review is an alinaqi/claude-bootstrap skill that delegates specialized code review to OpenAI's Codex CLI using GPT-5.2-Codex, a model trained for bug detection, security flaws, and quality issues. The SKILL.md positions it for multi-engine review alongside Claude Code sessions and documents CI/CD integration paths referencing the Codex CLI and OpenAI code review cookbook. Developers invoke it when requesting Codex-powered review or comparing findings across review engines on bootstrap repos. The skill is user-invocable with medium effort, making it a deliberate pre-merge gate rather than ambient linting. Reach for codex-review when a PR needs a second opinion on logic regressions, security gaps, or style drift that a primary agent might miss. It complements—not replaces—existing test suites and static analyzers by applying LLM reasoning to full diff context.289installs28Team Coordinationteam-coordination is an alinaqi/claude-bootstrap skill enabling multiple Claude Code sessions across a development team to work one repository without stepping on each other. The SKILL.md manages shared state, todo claiming, decision syncing, and session awareness so parallel agents know what is in progress, completed, or blocked. It activates when multiple developers share a repo and need bootstrap-friendly planning rituals plus status updates rather than ad-hoc chat coordination. The skill is not user-invocable by default and carries low effort, meaning agents should apply it automatically when multi-session conflict risk appears. Reach for team-coordination when handoffs, owner assignment, or milestone alignment across Claude Code terminals becomes the bottleneck. It formalizes who owns which task and which decisions are current, reducing duplicate work and conflicting edits during concurrent feature development.282installs29Typescripttypescript is a Claude skill from alinaqi/claude-bootstrap enforcing non-negotiable strict mode across **/*.ts, **/*.tsx, and tsconfig*.json files. The skill mandates compilerOptions including strict, noImplicitAny, strictNullChecks, noUnusedLocals, noUnusedParameters, and noImplicitReturns in tsconfig.json. Developers reach for typescript when editing React components or Node modules in the bootstrap monorepo and need consistent interface, generic, and module boundary patterns paired with ESLint and Jest. The skill defines a src/core pure-business-logic project structure for medium-effort TypeScript work. It auto-triggers on TypeScript file paths rather than user invocation.280installs30Reddit Apireddit-api is a Claude skill from alinaqi/claude-bootstrap for integrating Reddit data into applications via official API endpoints. The skill covers OAuth2 app creation at reddit.com/prefs/apps, script and web app types, and client libraries PRAW for Python and Snoowrap for Node.js. Developers use reddit-api to fetch posts, comments, subreddits, and user data, plus post and moderate content through authenticated endpoints. Reach for reddit-api when building Reddit bots, community research agents, or social distribution automations that need documented OAuth2 flows. The skill references official Reddit API docs, OAuth2 wiki, and PRAW documentation for medium-effort integrations.278installs31Cloudflare D1cloudflare-d1 teaches Claude how to use Cloudflare D1 as edge SQLite storage: schema migrations, Wrangler bindings, queries from Workers, and local development parity. Use it when building serverless APIs or SaaS features that need relational data close to Cloudflare edge compute without managing Aurora-scale infrastructure.272installs32Database SchemaGuides Claude through designing production-ready database schemas including entity relationships, normalization choices, primary and foreign keys, indexes, and migration strategies aligned with claude-bootstrap project conventions.272installs33Ms Teams AppsGuides construction of Microsoft Teams applications within claude-bootstrap: register the app, implement bot and tab surfaces, secure Graph access, and connect backend services. Targets B2B SaaS teams embedding workflows in Teams rather than building a standalone consumer mobile client.269installs34Session Managementsession-management is a Claude Code skill from alinaqi/claude-bootstrap for maintaining context across long development sessions and enabling seamless resume after breaks. The core principle is checkpoint at natural breakpoints and resume instantly. Three tiered summarization levels govern updates: Tier 1 quick updates to current-state.md after small tasks, Tier 2 full checkpoints adding decisions.md after significant changes or roughly 20 tool calls, and Tier 3 session archives under archive/YYYY-MM-DD.md after major milestones or about 50 tool calls. The skill defines a _project_specs/session/ directory with current-state.md, decisions.md, code-landmarks.md, and dated archive files. Developers reach for session-management when multi-hour Claude Code work risks losing architectural decisions, active file context, or next-step clarity across session boundaries. Session start protocol requires reading current-state.md and announcing resume point; session end protocol creates archives with specific handoff next steps. The skill is user-invocable false with low effort, designed as passive enforcement via CLAUDE.md session rules.267installs35Ai ModelsAI models skill helps teams compare LLM providers and tiers, mapping reasoning, multimodal, and tool-use strengths to product requirements during scoping so agent and API features are built on realistic cost, latency, and capability assumptions.266installs36Llm PatternsThe llm-patterns skill from alinaqi/claude-bootstrap documents practical patterns for prompts, tool use, retrieval, and failure handling so bootstrap projects embed LLM and agent behavior consistently instead of reinventing fragile one-off integrations.262installs37Web PaymentsCovers implementing web payment systems with provider SDKs, secure checkout session creation, subscription and invoice management, webhook event processing, refund handling, and error recovery patterns for SaaS and ecommerce products.262installs38Firebasefirebase guides Claude through Google Firebase integration for mobile and web products: Authentication, Firestore rules and modeling, Cloud Storage, Cloud Functions, and Hosting deployment. Use it when shipping MVPs or production apps that lean on managed realtime data, push-ready backends, and rapid serverless feature delivery.261installs39Agent TeamsThe agent-teams skill from alinaqi/claude-bootstrap teaches how to structure multiple Claude agents with clear roles, delegation rules, and handoff conventions so complex projects run as coordinated teams instead of one overloaded assistant.259installs40Supabase PythonBootstrap Python backends on Supabase: configure clients, run migrations, enforce row-level security, query Postgres, and use Auth, Storage, and Realtime so SaaS and agent apps get a production-ready data and identity layer quickly.259installs41Existing RepoBootstraps Claude agent tooling for existing repositories by discovering layout, documenting build and test commands, aligning with project conventions, and installing context files so assistants code productively in non-greenfield codebases.251installs42Supabase NodeIntegration skill for Supabase with Node.js in claude-bootstrap: client initialization, auth flows, Postgres access, row-level security, storage, and realtime patterns for SaaS and agent backends.251installs43CredentialsThe credentials skill from alinaqi/claude-bootstrap helps teams store and use API keys, tokens, and environment secrets safely in bootstrap projects, reducing leak risk and establishing rotation-friendly patterns before code reaches staging or production.250installs44Aws Auroraaws-aurora guides Claude on Amazon Aurora setup for SaaS and API backends: engine choice, HA failover, VPC security, connection pooling, backups, migrations, and read-replica scaling. Use it when replacing self-managed Postgres/MySQL or hardening a production relational tier on AWS.248installs45MedusaThe medusa skill from alinaqi/claude-bootstrap helps developers stand up and extend a Medusa headless commerce backend in bootstrap projects, covering core commerce entities, API routes, plugins, and integration patterns for modern ecommerce applications.247installs46Reddit AdsMarketing skill for Reddit advertising within the bootstrap kit: audience targeting, ad creative, campaign structure, budget framing, and landing alignment to acquire users at launch for SaaS, ecommerce, and content products.246installs47Iterative DevelopmentTeaches Claude Code to ship software in tight iterative loops: define a minimal vertical slice, implement it, run checks, capture learnings, and repeat. Suited to alinaqi/claude-bootstrap projects where agent sessions must stay focused, reversible, and continuously validated rather than sprawling refactors.245installs48Gemini ReviewThe gemini-review skill from alinaqi/claude-bootstrap introduces a Gemini-powered review workflow for bootstrap code changes, giving teams a cross-model quality gate that challenges Claude-generated patches and surfaces issues before merge.242installs49Ticket CraftSkill for authoring high-quality engineering tickets in bootstrap workflows: clear problem statements, bounded scope, verification steps, and links to site-architecture or integration specs. Makes agent sessions and human reviewers align on done-ness, reducing rework across frontend, backend, and integration streams.242installs50Project ToolingDefines how alinaqi/claude-bootstrap repos set up automation: package scripts, formatting, testing entrypoints, and conventions agents can invoke reliably. Reduces one-off setup drift so iterative development, reviews, and integrations share a common, documented toolchain from day one of the build.240installs51PythonProvides Claude Code conventions for writing clean Python applications including virtual environments, module structure, typing, error handling, packaging, and test patterns consistent with alinaqi claude-bootstrap workflows.240installs52Basebase from claude-bootstrap provides foundational project scaffolding for Claude Code workflows: directory layout, CONTEXT.md patterns, agent rules, and kickoff conventions. Use it when starting a new codebase or agent-assisted product so structure, documentation, and tooling defaults are set before feature work.237installs53Azure CosmosdbThe azure-cosmosdb skill from alinaqi/claude-bootstrap helps developers wire Azure Cosmos DB into bootstrap applications, covering SDK integration, schema and partition design, query patterns, and environment setup for scalable document storage.231installs54Shopify AppsTeaches Claude how to build Shopify apps end-to-end including partner dashboard setup, OAuth installation flow, webhook event handlers, Admin API usage, app extensions, and billing patterns from claude-bootstrap templates.228installs55Code GraphCode-graph from claude-bootstrap builds a navigable map of modules, imports, and relationships across a repository, giving agents and developers structural context for safer refactors, reviews, and targeted changes in large codebases.219installs56WorkspaceThe workspace skill from alinaqi/claude-bootstrap initializes a Claude Code-ready project workspace with standard directories, configuration stubs, and conventions so agents can navigate and contribute immediately without manual setup.158installs57Cpg Analysiscpg-analysis performs deep code analysis beyond the AST using Joern for a full code property graph and CodeQL for interprocedural taint analysis and vulnerability detection. It maps control flow, data flow, and program dependencies, and traces tainted data from source to sink for security auditing. A developer uses it for control flow, data flow, dead code, and complex refactoring analysis, or for a security audit before release. It is opt-in and requires Docker, a JVM, or the CodeQL CLI.132installs58Icpgicpg adds an intent-augmented code property graph, a Reason Graph layer on top of code structure that links every function and module to the goal, owner, and contract that created it. Before writing code, an agent runs three canonical queries for prior work, constraints, and symbol risk, and it detects when code drifts from its stated purpose across six dimensions. A developer uses it to avoid duplicate work, preserve invariants, and audit intent during autonomous development. State lives in a per-project SQLite database.118installs59Mnemosmnemos is a task-scoped memory lifecycle that prevents lossy context compaction from destroying the knowledge an agent needs. It stores working memory as a typed graph where goals and constraints are never evicted, results are compressed before eviction, and checkpoints persist to disk for session resume. It monitors agent fatigue and auto-checkpoints before compaction, then re-injects the checkpoint afterward. A developer uses it to preserve decisions and task handoffs across compactions.96installs60Maggymaggy is a local AI engineering command center that triages tickets and spawns Claude Code runs against any repo. It ranks open issues by urgency, OKR alignment, and recency, offers one-click Execute that launches Claude Code with iCPG context injected to run a TDD pipeline, and generates a daily competitor intelligence briefing. A developer uses it as a persistent dashboard for morning triage and ticket implementation across multiple repos and issue trackers.93installs61Cross Agent Delegationcross-agent-delegation lets Claude Code orchestrate task routing to other AI CLI tools like Kimi and Codex behind the scenes. It scores a task on five complexity dimensions, sourced from iCPG signals and reasoning, then routes low-risk work to Kimi, medium-risk work to Kimi plus Codex auto-review, and high-risk work to Claude directly. A developer uses it to delegate bounded work safely while keeping security-critical changes in Claude's context.83installs62Polyphonypolyphony is a multi-agent orchestration skill that runs each agent session in its own Docker container with a full git clone on its own branch. A supervisor loop discovers tasks from GitHub Issues or a local queue, routes each by a five-dimension complexity score to Claude, Codex, or Kimi, provisions a container, runs it, verifies, and lands the result. A developer uses it to run parallel agents without conflicts, with independent tests and clean per-branch PRs.77installs63Council Reviewcouncil-review provides behavioral rules for a multi-model validation council that reviews plans, architecture changes, and PRs before execution. When a plan is written or an architectural change is made, it runs validate-plan or review across a panel of models led by a chief model, then applies threshold voting to decide whether to auto-execute or surface feedback. A developer relies on it so architecture and security decisions get independent verification before they run.43installs